About 70% of neonatal deaths occur in premature infants. Therefore, since mortality during the first year of life is increasingly concentrated in the neonatal period, reduction of infant mortality has become primarily a problem of keeping premature infants alive. By far the largest factor in the deaths of such infants is impairment of pulmonary ventilation.1 Thus it is not surprising that additional atmospheric oxygen has been used increasingly in nurseries for premature infants.
Recent and rapidly growing literature2 reports a significant association between prolonged or concentrated oxygen administration and the retinal vascular changes that may culminate in the membrane formation described by Terry in 1942 as retrolental fibroplasia. On July 1, 1953, a group of investigators from 18 hospitals in various parts of the United States embarked on a cooperative study of this problem, supported by the National Institute of Neurological Disease and Blindness of the U. S.